I'm just starting Scanlon's "Reasons: A Puzzling Duality?", in Reason and Value: Themes from the Moral Philosophy of Joseph Raz. He considers the question of whether an agent's intended ends can give her reasons that she otherwise would not have had. This sounds plausible to me. But, there is supposed to be a puzzle if we also want to accept the claim that we can have reasons which are independent of our choices. Here is Scanlon:
"This gives rise to a puzzle about what it is to be a reason - that is to say, a consideration that counts in favor of some action or attitude. If the normative status of counting in favor of acting a certain way is something that certain considerations can just have, how can it also be something that we can confer on certain considerations by our choice of ends?" (p. 231)
Now, I think that Scanlon will go on to argue that there is not really a puzzle here, after all. But, I just don't feel even the initial force. Am I missing something? Why is this combination supposed to be puzzling?
Tuesday, May 26, 2009
Monday, May 18, 2009
Reading list
I'm trying to post more often. So, in that spirit, here's my reading list for the summer.
1. Scanlon's Moral Dimensions: Permissibility, Meaning, Blame (for reading group)
2. A bunch of stuff on contrastivism (for a reading group)
3. Cappelen & Hawthorne's Relativism and Monadic Truth
4. A biography of Tarski by Solomon Feferman and Anita Burdman Feferman that I found on half.com for like four dollars. I love half.com.
5. The manuscript for Mark Schroeder's new book, Noncognitivism in Ethics
That's my ambitious list. I may not make it through everything, but I'll be working in the philosophy library four days a week, so I'll have a lot of time to myself.
Others?
1. Scanlon's Moral Dimensions: Permissibility, Meaning, Blame (for reading group)
2. A bunch of stuff on contrastivism (for a reading group)
3. Cappelen & Hawthorne's Relativism and Monadic Truth
4. A biography of Tarski by Solomon Feferman and Anita Burdman Feferman that I found on half.com for like four dollars. I love half.com.
5. The manuscript for Mark Schroeder's new book, Noncognitivism in Ethics
That's my ambitious list. I may not make it through everything, but I'll be working in the philosophy library four days a week, so I'll have a lot of time to myself.
Others?
Thursday, May 14, 2009
Are all reasons contrastive?
I'm reading through chapter 5 of Sinnott-Armstrong's Moral Skepticisms, where he lays out his contrastive framework for justified belief. Near the beginning of the chapter, he has a few paragraphs about contrastive reasons. Since I'm interested to see what applications contrastivism can have in metaethics, I found this (and the rest of the chapter, of course) pretty interesting.
Some reasons are contrastive. A can be a reason for B rather than C, but maybe not for B rather than D. For example, that it is my friend's birthday is a reason for me to bake him a chocolate cake rather than no cake at all, but it may not be a reason for me to bake him a chocolate cake rather than a vanilla cake. Reasons for that would come from things like my friend's preferences.
That it is a nice sunny southern California day is a reason for me to go to the beach rather than staying inside, but it may not be a reason for me to go to the beach rather than going for a hike. A reason for this would need to come from some other factors.
So, some reasons for action are contrastive. Similarly, some moral reasons seem to be contrastive. That my friend is sick in the hospital is a reason to send him a card rather than ignoring him, but it may not be a reason for me to send him a card rather than going to visit him.
After surveying similar examples, Sinnott-Armstrong hypothesizes that all reasons are contrastive, including reasons for belief, or justificatory reasons. If I see a security video with someone who looks the same as my twin sisters look stealing a car, then I have reason to believe that my sister Amanda stole the car rather than that my friend Chris stole the car, but I may not have a reason to believe that my sister Amanda stole the car rather than that my sister Kelli stole the car.
I think much of this is right. But, at least at first, I was a bit resistant. After all, it seems that the fact that it is my friend's birthday gives me a reason to bake him a chocolate cake simpliciter. I may have much more reason to bake him a vanilla cake, if he prefers vanilla, but don't I still have some reason simpliciter to bake him a chocolate cake, assuming he doesn't hate chocolate, isn't allergic, etc.? And if not - because, for example, he hates chocolate - then does the fact that it is his birthday even give me a reason to bake him a chocolate cake?
Doesn't the fact that I see someone who looks like my twin sisters give me some reason simpliciter for thinking that Amanda stole the car? I may not have enough of a reason to rule out the competing hypothesis that Kelli stole the car, but don't I still have some reason to think Amanda did?
But, the contrastivist definitely has a response. Yes, that the person in the security tape looks like my twin sisters gives me some reason to think that Amanda stole the car. But, this reason (like all reasons, perhaps?) contrastive. It is perfectly legitimate to say simply that I have a reason to think Amanda stole the car. But, this is because I have reason to think Amanda stole it rather than that (say) Chris stole it. But, it is intuitively wrong to say that I have a reason to think Amanda stole the car rather than Kelli (given the right specifications of the case). There just is no such thing as a reason simpliciter.
What do people think about contrastive reasons? Are all reasons contrastive?
Some reasons are contrastive. A can be a reason for B rather than C, but maybe not for B rather than D. For example, that it is my friend's birthday is a reason for me to bake him a chocolate cake rather than no cake at all, but it may not be a reason for me to bake him a chocolate cake rather than a vanilla cake. Reasons for that would come from things like my friend's preferences.
That it is a nice sunny southern California day is a reason for me to go to the beach rather than staying inside, but it may not be a reason for me to go to the beach rather than going for a hike. A reason for this would need to come from some other factors.
So, some reasons for action are contrastive. Similarly, some moral reasons seem to be contrastive. That my friend is sick in the hospital is a reason to send him a card rather than ignoring him, but it may not be a reason for me to send him a card rather than going to visit him.
After surveying similar examples, Sinnott-Armstrong hypothesizes that all reasons are contrastive, including reasons for belief, or justificatory reasons. If I see a security video with someone who looks the same as my twin sisters look stealing a car, then I have reason to believe that my sister Amanda stole the car rather than that my friend Chris stole the car, but I may not have a reason to believe that my sister Amanda stole the car rather than that my sister Kelli stole the car.
I think much of this is right. But, at least at first, I was a bit resistant. After all, it seems that the fact that it is my friend's birthday gives me a reason to bake him a chocolate cake simpliciter. I may have much more reason to bake him a vanilla cake, if he prefers vanilla, but don't I still have some reason simpliciter to bake him a chocolate cake, assuming he doesn't hate chocolate, isn't allergic, etc.? And if not - because, for example, he hates chocolate - then does the fact that it is his birthday even give me a reason to bake him a chocolate cake?
Doesn't the fact that I see someone who looks like my twin sisters give me some reason simpliciter for thinking that Amanda stole the car? I may not have enough of a reason to rule out the competing hypothesis that Kelli stole the car, but don't I still have some reason to think Amanda did?
But, the contrastivist definitely has a response. Yes, that the person in the security tape looks like my twin sisters gives me some reason to think that Amanda stole the car. But, this reason (like all reasons, perhaps?) contrastive. It is perfectly legitimate to say simply that I have a reason to think Amanda stole the car. But, this is because I have reason to think Amanda stole it rather than that (say) Chris stole it. But, it is intuitively wrong to say that I have a reason to think Amanda stole the car rather than Kelli (given the right specifications of the case). There just is no such thing as a reason simpliciter.
What do people think about contrastive reasons? Are all reasons contrastive?
Tuesday, May 12, 2009
Contrastivism reading group
I'm putting together a reading group for the summer on contrastivism, especially contrastivism about epistemology. We may get to a few things on contrastivism in metaethics (I'm particularly interested in exploring the prospects here), or contrastivism in causation.
I've found a bunch of readings, though I don't think we'll get through them. The plan is to start sometime near the beginning of June. I set up a drop/blog at
http://drop.io/usccontrastivism
We'll have weekly meetings at USC, but a nice feature of using drop.io is that we can easily share files and blog posts, with little write-ups about the discussion, so people from other places can join in (at least to some extent). If you're interested, let me know, and I'll put you on the email list.
I've found a bunch of readings, though I don't think we'll get through them. The plan is to start sometime near the beginning of June. I set up a drop/blog at
http://drop.io/usccontrastivism
We'll have weekly meetings at USC, but a nice feature of using drop.io is that we can easily share files and blog posts, with little write-ups about the discussion, so people from other places can join in (at least to some extent). If you're interested, let me know, and I'll put you on the email list.
Monday, May 11, 2009
What do we want from an account of evidence?
I've just finished Ram Neta's paper, "What Evidence Do You Have?", and I'm thinking about what evidence is for. Neta says that an account of evidence must allow for, and explain, these two adequacy conditions.
(i) Particular bits of evidence (which Neta (as well as Williamson, and many other philosophers, I'm sure) take to be propositions) can be used to rationally regulate our degrees of confidence in various hypotheses.
(ii) Epistemic rationality requires that, for any hypothesis h, your confidence in h is proportional to the support that h receives from your total evidence.
The question, as Neta poses it, then, is "Why is it that your evidence has these two epistemic powers?". Neta argues that the existing accounts of evidence (e.g. E=K, Lewis's account from "Elusive Knowledge", etc.) cannot answer this question.
I'm interested to know if people generally agree on these adequacy conditions, or if they think one, or both, should be rejected, or if there should be more, or whatever.
In particular, I'm interested in seeing if E=B, that is, that one's evidence consists of all and only one's beliefs, can be defended. If the above two adequacy conditions are right, I think that lends some support to E=B. Plausibly, rationality requires us to have consistent beliefs. This requirement would seem to generalize to conditions like (i) and (ii) - our degrees of confidence in various propositions should be proportional to the support they receive from our beliefs.
Neta argues against E=B with something like the following case:
I know that someone is drawing balls from an urn, and that they have drawn 99 balls so far. But, I haven't been paying any attention to what color the balls drawn so far have been. But, nevertheless, I come to believe that all 99 balls drawn so far have been black. Then, if E=B, one of the propositions in my total evidence is that the first 99 balls drawn have been black. Then, it seems that I should be very confident in the proposition that the 100th ball drawn will be black. But, this can't be right. So, E=B is false.
I did not, for what it's worth, find this case convincing. If I really believe that the first 99 balls drawn are black (whether or not this belief is justified), doesn't it seem like I should raise my confidence in the truth of the proposition that the 100th ball drawn will be black?
Anyway, E=B does seem problematic. But, I am interested to see if it is more defensible than Neta seems to suggest. Any suggestions about what to look at on that topic would be appreciated.
(i) Particular bits of evidence (which Neta (as well as Williamson, and many other philosophers, I'm sure) take to be propositions) can be used to rationally regulate our degrees of confidence in various hypotheses.
(ii) Epistemic rationality requires that, for any hypothesis h, your confidence in h is proportional to the support that h receives from your total evidence.
The question, as Neta poses it, then, is "Why is it that your evidence has these two epistemic powers?". Neta argues that the existing accounts of evidence (e.g. E=K, Lewis's account from "Elusive Knowledge", etc.) cannot answer this question.
I'm interested to know if people generally agree on these adequacy conditions, or if they think one, or both, should be rejected, or if there should be more, or whatever.
In particular, I'm interested in seeing if E=B, that is, that one's evidence consists of all and only one's beliefs, can be defended. If the above two adequacy conditions are right, I think that lends some support to E=B. Plausibly, rationality requires us to have consistent beliefs. This requirement would seem to generalize to conditions like (i) and (ii) - our degrees of confidence in various propositions should be proportional to the support they receive from our beliefs.
Neta argues against E=B with something like the following case:
I know that someone is drawing balls from an urn, and that they have drawn 99 balls so far. But, I haven't been paying any attention to what color the balls drawn so far have been. But, nevertheless, I come to believe that all 99 balls drawn so far have been black. Then, if E=B, one of the propositions in my total evidence is that the first 99 balls drawn have been black. Then, it seems that I should be very confident in the proposition that the 100th ball drawn will be black. But, this can't be right. So, E=B is false.
I did not, for what it's worth, find this case convincing. If I really believe that the first 99 balls drawn are black (whether or not this belief is justified), doesn't it seem like I should raise my confidence in the truth of the proposition that the 100th ball drawn will be black?
Anyway, E=B does seem problematic. But, I am interested to see if it is more defensible than Neta seems to suggest. Any suggestions about what to look at on that topic would be appreciated.
Sunday, April 19, 2009
Linguistic competence
We've been reading through Williamson's The Philosophy of Philosophy in my writing seminar, and focusing on Williamson's arguments that linguistic competence cannot play any substantial epistemological role. So, naturally, I've been thinking about linguistic competence.
I found Williamson's examples of competent speakers who reject seemingly obvious, analytic truths interesting. For example, he invents two theorists (based on real-life examples, I'm sure) who each reject (for different reasons) the sentence "Every vixen is a vixen". The first, Peter, is convinced that the sentence can only be true if the sentence "There is at least one vixen" is true (he thinks the universally quantified sentence entails the existentially quantified sentence), and that there are no vixens (it's all a conspiracy). The second, Stephen, worries about vagueness. At some point in the evolutionary history of foxes, there were borderline cases - we can't really say that the individuals are foxes, nor that they are not foxes. So, we can't really say that the females are vixens, nor that they are not vixens. So, the universally quantified sentence is indeterminate - we should not assent to it, nor should we assent to its negation (this example is based on someone with Scott's views on vague predicates).
Anyway, the arguments (and Scott's writing assignments) got me trying to think of sentences that competent speakers cannot reject. It isn't so easy - which is why Williamson draws the strong conclusion that there are none.
But, even if that strong conclusion is right, what about this. Consider the following schema:
(T) It is true that S iff S.
Liar paradox worries may stop speakers from assenting to every instance of (T) (that they understand), but shouldn't they assent to all nonparadoxical instances? Maybe not. Non-factualists about some area of discourse may assent to the right hand side ("That is a beautiful painting"), but refuse to assent to the left hand side ("It is true that that is a beautiful painting"). So, something weaker: shouldn't competent speakers at least not routinely assent to the negations of instances of (T)?
I found Williamson's examples of competent speakers who reject seemingly obvious, analytic truths interesting. For example, he invents two theorists (based on real-life examples, I'm sure) who each reject (for different reasons) the sentence "Every vixen is a vixen". The first, Peter, is convinced that the sentence can only be true if the sentence "There is at least one vixen" is true (he thinks the universally quantified sentence entails the existentially quantified sentence), and that there are no vixens (it's all a conspiracy). The second, Stephen, worries about vagueness. At some point in the evolutionary history of foxes, there were borderline cases - we can't really say that the individuals are foxes, nor that they are not foxes. So, we can't really say that the females are vixens, nor that they are not vixens. So, the universally quantified sentence is indeterminate - we should not assent to it, nor should we assent to its negation (this example is based on someone with Scott's views on vague predicates).
Anyway, the arguments (and Scott's writing assignments) got me trying to think of sentences that competent speakers cannot reject. It isn't so easy - which is why Williamson draws the strong conclusion that there are none.
But, even if that strong conclusion is right, what about this. Consider the following schema:
(T) It is true that S iff S.
Liar paradox worries may stop speakers from assenting to every instance of (T) (that they understand), but shouldn't they assent to all nonparadoxical instances? Maybe not. Non-factualists about some area of discourse may assent to the right hand side ("That is a beautiful painting"), but refuse to assent to the left hand side ("It is true that that is a beautiful painting"). So, something weaker: shouldn't competent speakers at least not routinely assent to the negations of instances of (T)?
End of the semester
The spring semester is wrapping up, so I'm working on papers for my seminars. For my epistemology seminar, I'm working on a paper presenting some problems for E=K. I don't have all the arguments worked out yet, but I think there are some worth making. For the metaethics seminar, I'm working on a paper surveying the literature, and suggesting some possibly fruitful areas of future research, concerning the wide-scope view of `ought' and how oughts and reasons transmit from ends to means. I also have a short paper in the works on Davidson's interpretation of Tarski in Truth and Predication that started as an assignment for my Writing for Publication seminar.
This summer, besides the Scanlon reading group and the contrastivism reading group, I'm going to read through the manuscript for Mark Schroeder's Noncognitivism in Ethics. I'm interested in exploring some issues in moral semantics, and this book will provide some good background, I think.
This summer, besides the Scanlon reading group and the contrastivism reading group, I'm going to read through the manuscript for Mark Schroeder's Noncognitivism in Ethics. I'm interested in exploring some issues in moral semantics, and this book will provide some good background, I think.
Subscribe to:
Posts (Atom)
